The first detailed accounts of Poulnagollum and Upper Poulelva are given in papers by J. C. Coleman and N. J. Dunnington in 1944 and 1949· Many of the cave passages were named in those accounts and as far as possible the same names have been used in this paper. The explanation for the names is usually self evident and so not all the explanations are given here. Further, for the sake of brevity, the papers of Coleman and Dunnington are hereafter cited as “C. & D. 1944 and 1949” respectively. Since these accounts appeared a good many new discoveries have been made greatly extending the system and proving various connexions. In the course of surveying these new passages still more were found and it became necessary to do a check survey, originally without passage detail, of the parts surveyed by C. & D. so as to be able to tie in the new passages to their survey. In the course of this work various discrepancies between our survey results and the earlier accounts were found. Some of the discrepancies were of a major nature and seem likely to have been due to errors of computation, e.g., adding lengths together twice. Because of these errors it was decided to make a completely new survey. This has given a most interesting plan of the layout of the passages in relation to each other. These passages, with two exceptions, are all described from their explorable beginning in a downstream direction so that all lefts and rights are used in the same sense. For Branch Passage and the Craven Canyon this arrangement is not practicable and these two are described moving away from the point of entry, that is upstream in Branch Passage. The method followed in this account is to describe the main route from the head of the Short Gallery of Upper PoulnagolJum through the cave to the Flooded Bedding Plane Passage and the Maze. Although the streamway in Lower Poulelva is a continuation of that in Poulnagollum, it was found convenient to describe the whole of the Poulelva series separately. The tributaries are described in groups at the end of each account of a major section of the system. In the Bibliography a selection has been made of the principal references. Thus references in caving club journals about visits to the cave are omitted as they do not provide any additional information. The spelling of names is taken from the earliest sources. Some are taken from the Ordnance Survey maps and most of the rest from the papers of C. & D. (1944 and 1949).